From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!wupost!uunet!psinntp!scylla!daryl Tue Feb 11 15:25:53 EST 1992
Article 3597 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!wupost!uunet!psinntp!scylla!daryl
>From: daryl@oracorp.com
Newsgroups: comp.ai.philosophy
Subject: Re: Strong AI and panpsychism
Message-ID: <1992Feb7.154307.2106@oracorp.com>
Date: 7 Feb 92 15:43:07 GMT
Organization: ORA Corporation
Lines: 81

David Chalmers writes: (In response to Michael Gemar's comments
about conscious rocks:)

> Still, Putnam doesn't actually need time-varying states for the result
> to go through, on the assumption that any object has an (implicit or
> explicit) "clock" built into it, that ensures that it is in different
> states at different times. But the states that appear are
> nevertheless highly gerrymandered.

Being "highly gerrymandered" seems to me to be a matter of the
difficulty of interpreting a rock as conscious, but it doesn't seem to
be relevant to the question of whether it can be so interpreted. Why
should a convoluted, complex definition of the mental states count any
less than a straight-forward one?

I am not sure what notion of state machine that Putnam was using for
his proof that a physical object simultaneously implements every FSA,
but let me discuss this from the point of view of process algebras.
(for example, Robin Milner's CCS (Calculus of Communicating Systems)).

In old-fashioned Finite Automaton theory (for example, see Hopcroft
and Ulman), a finite state machine is determined by a set of states,
an alphabet of visible signals, and a transition relation between
states. In process algebra, however, two states are only considered
different if they *behave* differently; if they have different
transitions. This can be made precise by starting with a finite
state machine and identifying isomorphic states.

The problem in applying this theory to a physical object is that there
is no immediate interpretation of what the alphabet (set of possible
signals) is. When do you consider the physical object to be "giving
out different signals"? If you confine your attention to a particular
mode of interaction with the object (for example, in the Turing Test,
the mode of interaction is restricted to an exchange of characters
appearing on the keyboard), then you can use this as the definition of
the alphabet, and the theory of processes uniquely determines what
process the physical object implements.

This approach is unsatisfactory; a being may be conscious, in the
intuitive sense, yet may be incapable of producing keyboard
characters, for physical reasons or because lack of familiarity with
keyboards. That's fine, so you choose some other set of signals;
perhaps you use a set of sounds, or perhaps you actually use the
electrical signals appearing in the being's brain as the set of
signals.

However, the process that a physical object implements is not a
function solely of the physical laws governing the dynamics of that
object---it also depends crucially on the choice of physical
circumstances that are to be considered the signals. This relates to
my previous discussion with Michael Gemar about whether being
conscious is an objective fact or not. I believe that for a given
choice of a set of signals, it may be an objective fact that an object
is conscious. The choice of what set of signals are "reasonable" or
"sensible" or "natural" is not, in my opinion, objective.

If you ask the question "Is there *some* set of signals for which a
rock can be considered to implement a conscious process?" then I am
willing to believe that the answer is yes. If that set of signals is
not in any way considered "sensible" to us, then we would not consider
the rock conscious. The rock may not consider us conscious, either.

Michael Gemar's, and Jeff Dalton's, and David Gudeman's insistence
that it is a fact, rather than a matter of opinion, that they are
conscious is both obviously true and obviously false, according to the
process notion I have been discussing. When I say "I am conscious",
the question is what the "I" refers to. If it refers to the physical
object that people sometimes call "Daryl McCullough", then I would say
that I have no introspective, first-hand knowledge that that body is
conscious, since I only have inconclusive circumstantial evidence that
the consciousness that I experience first-hand is *caused* by that
body. On the other hand, if the "I" refers to the internal process
that is being experienced first-hand (that *is* first-hand
experience), then it seems to me that this is an objective fact,
although whether that process is somehow implemented by a particular
body is not necessarily objective (given the problem in objectively
definining a meaningful set of signals for a physical object).

Daryl McCullough
ORA Corp.
Ithaca, NY


